1,4-Butanediol (BDO) is an important commodity chemical used in the manufacture of tetrahydrofuran, polyesters and polyurethanes. Most of the BDO produced commercially is made by the reaction of acetylene with two equivalents of formaldehyde followed by hydrogenation of the resulting alkyne. This process has various disadvantages including the use of relatively expensive and hazardous raw materials.
Some BDO is also produced by the reaction of acetic acid, oxygen, and 1,3-butadiene to produce 1,4-diacetoxy-2-butene which is then hydrogenated and hydrolyzed. This process suffers from various drawbacks including the number of steps involved and the co-production of 3-butene-1,2-diol diacetate.
Other processes for producing BDO include the hydrogenation of maleic anhydride, and the hydroformylation of allyl alcohol followed by hydrogenation of the intermediate 4-hydroxybutyraldehyde. These processes also suffer from various drawbacks such as requiring severe operating conditions and requiring expensive rhodium catalyst, respectively.
An attractive route to BDO is through the hydrogenation of 2-butene-1,4-diol (hereinafter "1,4-butenediol"). However, there is no known way to synthesize 1,4-butenediol safely, efficiently, and inexpensively. It is known that 3,4-epoxy-1-butene (hereinafter "EPB") can be made efficiently from 1,3-butadiene and oxygen (see, e.g., U.S. Pat. Nos. 4,897,498 and 4,950,773), but there is no known process that can hydrolyze EPB to 1,4-butenediol with sufficiently high yield and selectivity.
For example, in J. Am. Chem. Soc., 104, 1658-1665 (1982), Ross et al. teach that acid-catalyzed hydrolysis of EPB produces a mixture containing 96% 3-butene-1,2-diol (hereinafter "1,2-butenediol") and only 4% 1,4-butenediol. Likewise, we have found that hydrolysis of EPB with aqueous sodium hydroxide gives 1,2-butenediol, high boilers, and little or no 1,4-butenediol. Thus, neither acid nor base catalysis conditions are suitable for the hydrolysis of EPB to a product containing useful levels of the desired 1,4-butenediol isomer.
Japanese Kokai Patent No. 54-79214 describes a process that uses hydriodic acid and, optionally, a transition metal compound as a catalyst for the hydrolysis of EPB to mixtures containing 1,4-butenediol. Under the most selective conditions reported, the diol mixture had a 1,4/1,2 ratio of only 1.3 and a total diol yield of only 59%. Thus, this process not only gives a poor yield, but is also corrosive. Additionally, no method for the separation of the diol products from the catalyst components is disclosed.
Japanese Kokai Patent No. 54-73710 describes the use of both Cu(I) and Cu(II) salts as catalysts for the hydrolysis of EPB to mixtures of 1,4-butenediol and 1,2- butenediol. However, the reactions shown in the examples thereof were very slow and exhibited poor selectivity to the desired 1,4-isomer. For example, after 50 hours at elevated temperature, CuBr provided a mixture of dials having a 1,4/1,2 isomer ratio of only 0.34. Moreover, there is no disclosure of a process for separating and recovering the catalyst from the reaction product mixture.
U.S. Pat. No. 5,530,167 discloses a process for the hydrolysis of EPB to form a mixture of dials using a supported-copper catalyst. In an example thereof, it is reported that a NaY zeolite supported copper(II) catalyst gave a 1,4/1,2 ratio of 0.93.
Japanese Kokai Patent No. 57-2227 discloses a process for the hydrolysis of EPB to dials in the presence of an alkali metal iodide, alkaline-earth metal iodide, or zinc iodide and an acid selected from sulfuric acid, hydrochloric acid, hydrobromic acid, hydriodic acid, phosphoric acid and sulfonic acid. Under the best conditions reported, the 1,4/1,2 ratio was 5.15. Again, this process is highly corrosive and no method is disclosed for the separation of the products from the catalyst components.
Other known processes for the preparation of mixtures of 1,4-butenediol and 1,2-butenediol from EPB provide very low 1,4/1,2 ratios. For example, it is disclosed in DE 4429700 that EPB is hydrolyzed in the presence of rhenium oxide to give 3% 1,4-butenediol and 65% 1,2-butenediol (1,4/1,2 ratio=0.05). DE 4429699 discloses the hydrolysis of EPB in the presence of an insoluble oxide catalyst (e.g., 59% Sio.sub.2, 38% TiO.sub.2 and 0.25% F) to give 7% 1,4-butenediol and 54% 1,2-butenediol (1,4/1,2 ratio=0.13). And DE 4342030 discloses that non-catalyzed hydrolysis of EPB at 100.degree. C. gave 100% conversion to a product containing 14% 1,4-butenediol and 71% 1,2-butenediol (1,4/1,2 ratio=0.20).
From the above, it can be seen that non-iodide processes for the hydrolysis of EPB give very low 1,4/1,2 ratios. While known iodide processes provide relatively higher 1,4/1,2 ratios, the ratio is still not high enough and its variability is not low enough for the hydrogenation of 1,4-butenediol to be an attractive route to BDO. Moreover, such iodide processes are highly corrosive and, therefore, are not very attractive. Further, in the iodide processes, there is no disclosure of a method for separating the resulting products from the catalyst components. Thus, a need exists in the art for a process that can efficiently and selectively hydrolyze EPB to a product abundant in 1,4-butenediol with less corrosive effect. There is also a need in the art for a process for separating the resulting diol products from the catalyst components.
Accordingly, it is an object of the present invention to provide a process for the preparation of mixtures of 1,4-diol and 1,2-diol from .gamma.,.delta.-epoxyalkenes such as EPB having improved 1,4/1,2 selectivity and reduced variability.
It is a further object of the present invention to provide a process for the separation of catalyst components from the 1,4-diol and 1,2-diol product mixture.
These and other objects of the present invention will become apparent in light of the following specification, and the appended drawings and claims.